Depression is the most highly prevalent neuropsychiatric syndrome across all neurologic illnesses (Fig. 5-1). The relationship between depression and neurologic illness is complex in several aspects, including pathophysiology, clinical presentation, and response to treatment. Depression may sometimes be the direct result of brain pathology, very much like other neurological manifestations, such as cognitive or motor disturbances. In other cases, a more complex interplay of neurobiological, environmental, and coping mechanisms appears responsible for the development of depressive symptoms (Box 5-1). Neurologic illness may produce signs and symptoms that mimic depression, such as psychomotor retardation, apathy, concentration deficits, and sleep disorders, making the diagnosis of depression in the context of neurologic illness challenging. However, the identification and treatment of depression in these individuals is of fundamental importance, as this tends to improve the prognosis of the neurological disease.

This chapter covers depression in the context of specific neurological disorders, such as dementia, cerebrovascular disease, Parkinson disease (PD) and other movement disorders, multiple sclerosis, traumatic brain illness, and epilepsy. Depression in the context of pain is covered in Chapter 21 and depression in the context of a sleep disorder is covered in chapter 22.

See Figure 5-1 and Box 5-1 for prevalence and risk factors of depression in different neurological illnesses.

“Dementia” refers to a syndrome characterized by cognitive deterioration, behavioral abnormalities, and possible personality changes that significantly affect daily functioning. A dementia syndrome can be caused by myriad etiologies, including neurodegenerative processes, such as Alzheimer disease (AD), dementia with Lewy bodies, and frontotemporal dementia, cerebrovascular disease, traumatic brain injury (TBI), infectious illnesses such as HIV, and other neurologic disturbances. This section covers depression in dementia secondary to a neurodegenerative process, with emphasis on depression in AD. Depression refers to clinically significant depressive symptoms, rather than particular subtypes as defined by Diagnostic and Statistical Manual of Mental Disorders (DSM) unless otherwise noted. Cerebrovascular depression and depression associated with PD are addressed later in this chapter.

EPIDEMIOLOGY

Neuropsychiatric symptoms are common in dementia, and depression is one of the most frequent manifestations.^1,2 Much of the research in this area has focused on the relationship between depression and AD. The reported prevalence of depression in the AD population (dAD) varies widely. This heterogeneity is in part due to differences in depression criteria, depressive subtypes included, study settings, and assessment methods used in various studies.³ Estimates range between 1% and 90%, but the majority fall between 30% and 50%^3–5

There is very little published data on depressive subtypes within the dementia population. A 2010 study⁶ categorized patients with dementia into one of three groups by etiology: AD, vascular dementia, or unidentified dementia. Approximately 18.5% of patients with AD were depressed; 5.05% of that group met ICD-9 criteria for major depressive disorder (MDD), 1.93% for dysthymic disorder, 0.71% for adjustment disorder with depressive symptoms, and 0.20% depressive psychosis. An additional 12.8% were diagnosed with depressive disorder not otherwise specified.

PATHOPHYSIOLOGY

In AD, the pathophysiological process is thought to begin many years before the clinical symptoms. In an initial stage, there may be asymptomatic cerebral amyloidosis without evidence of neurodegeneration. Later on, markers of neuronal injury, such as elevated CSF tau, cortical thinning/gray matter loss, and cortical hypometabolism may be present. After the preclinical stage, often a period termed mild cognitive impairment (MCI) during which a person experiences a subjective sense of intellectual decline without a change in functional ability precedes diagnosis. Ample evidence suggests that depression and dementia are inextricably linked. Whether depression is a risk factor, prodrome, or consequence of AD is an ongoing area of research. Several studies suggest a history of depression earlier in life approximately doubles the risk of developing dementia.^7–11 On the other hand, late-life depression (usually defined as presenting at age 60 or older) may represent a prodrome of incipient dementia. Although a definitive pathophysiological link between depression and dementia has yet to be established, several theories have been proposed.^12,13 These include dysregulation of the hypothalamic–pituitary–adrenal (HPA) axis, sequelae of chronic inflammation, and nerve growth factor derangements. See Box 5-2 for possible mechanisms for depression in neurological illnesses.

DYSREGULATION OF THE HYPOTHALAMIC–PITUITARY–ADRENAL AXIS

Major depression is associated with elevated cortisol levels, mediated by increased cortisol release and lack of sensitivity to normal negative feedback mechanisms. Excess cortisol is linked to hippocampal atrophy in animal models and human studies, an anatomical change that also occurs in AD.¹⁴ In addition, longitudinal studies suggest that chronically elevated glucocorticoids levels can adversely affect memory.¹⁵ Glucocorticoids can trigger apoptotic death in hippocampal neurons via activation of the glucocorticoid receptors.¹⁶ In animal models of AD, stress level glucocorticoid administration promotes beta-amyloid deposition.¹⁷

CHRONIC INFLAMMATION

Both depression and cognitive decline may be mediated through inflammation. Depression is considered a proinflammatory state, with increased levels of cytokines, such as interleukin-6 (IL-6) and tumor necrosis factor (TNF). These cytokines can (through a series of intermediary steps) reduce synaptic plasticity and hippocampal neurogenesis. Proinflammatory cytokines can also promote neurodegeneration by inducing neuronal apoptosis.¹⁸ Amyloid-β (Aβ) can activate microglia to release proinflammatory cytokines. Microglia activation can be found in patients even at the earliest stages of cognitive decline, such as MCI.¹⁹

NERVE GROWTH FACTORS

Neurotrophins are essential in modulating synaptic plasticity and maintenance of neuronal health.²⁰ Levels of certain neurotrophins, such as brain-derived neurotrophic growth factor (BDNF) and transforming growth factor (TGF), are reduced in both dementia and depression. Impaired BNDF signaling occurs in animal models of stress-induced depression, humans with depression, and individuals with AD. BDNF is essential in regulating hippocampal plasticity, and therefore likely involved in maintaining cognition and memory.

It bears mentioning that simply being aware of cognitive deficits, or being diagnosed with a neurodegenerative disorder, do not seem to necessarily result in a “reactive” depression. Several studies showed no correlations between awareness of cognitive deficits or diagnosis of cognitive impairment and depression scores.^21–24

CLINICAL PRESENTATION

In general, depression in dementia presents similarly to major depression in the nondemented population. Regarding depression in the setting of earlier and milder cognitive deficits, the most common symptoms of depression in MCI may be excessive worrying, hopelessness, and crying spells.^25,26 See Box 5-3 for symptoms commonly associated with depression in the different neurological illnesses.

Depression in AD is notable for a higher frequency of motivational disturbances such as psychomotor slowing and fatigue, whereas geriatric depression without cognitive impairment has a higher frequency of mood symptoms, such as depressed mood, anxiety, suicidality, and sleep and appetitive disturbances.²⁷ Studies have also found that depression with AD is associated with higher rates of delusions and other psychotic symptoms than general geriatric depression.^27,28 A 2011 study of AD patients found that within those with depression, sadness, loss of interest, and agitation/retardation were the most common among the depressive symptoms.²⁹ Sadness, anxiety, suicidal thoughts, poor self-esteem, multiple physical complaints, and pessimism best distinguished the depressed patients from the nondepressed.

Depression is considered part of the behavioral and psychological symptoms of dementia (BPSD) spectrum, and is frequently found with other comorbid neuropsychiatric symptoms. These may include anxiety, apathy, psychotic symptoms, mood lability, agitation, or some combination thereof.³⁰ In a 2013 study, Van der Mussele et al.³¹ found that a number of behavioral symptoms such as delusions, hallucinations, psychosis, activity disturbances, aggressiveness, diurnal rhythm disturbances, affective disturbances, and anxiety/phobia were more prevalent in depressed AD patients than nondepressed AD patients.

COURSE AND NATURAL HISTORY

Longitudinal studies of depression in patients with dementia suggest that mood symptoms may have an episodic, rather than chronic, course and that in a significant percentage of these patients symptoms may improve or remit in subsequent assessments.^32–34 In each of these studies, patients with depression would have had access to, and in some cases received, antidepressant treatment. The prevalence of depression and associated complaints (sadness, suicidal ideation, low self-esteem, guilt, anxiety, crying, and hopelessness) in patients with Alzheimer dementia has been found to be highest in the moderately demented population, followed by the mild dementia population.³⁵ Severe dementia is associated with the lowest prevalence of depressive symptoms. As dementia progresses from moderate to severe stages, patients may lose the ability to experience and communicate their psychological state. In addition, they may display other behavioral symptoms such as agitation, aggression, or anxiety, interfering with the identification of depressive symptoms.

ASSESSMENT

While individuals with MCI and mild to moderate dementia may underestimate their cognitive deficits, they remain key informants on their subjective mood state.³⁶ Caregiver input is needed on standard measures of depression, which require recall of sleep and eating patterns and engagement in activities; caregivers also have important observations on affect and behavioral manifestations, such as tearfulness and withdrawal. Although caregivers’ reports may be influenced by their own level of distress, studies still show high concordance between clinician-rated, caregiver-observed, and patient-reported mood.³⁶ The best practice for assessment is a clinical interview with both patient and caregiver.

DIAGNOSTIC CRITERIA

There is substantial overlap between some of the depression criteria and symptoms of dementia (e.g., difficulties with concentration and memory, apathy, and fatigue). In 2001, using the existing criteria for major depression as laid out in the DSM-IV, an expert panel convened by the National Institute for Mental Health (NIMH) created a provisional set of diagnostic criteria for depression in patients with AD: the NIMH-dAD (Table 5-1).

The minimum number of symptoms required to diagnose depression in AD was decreased from 5 to 3 and the requirement that symptoms be present “most of the day, nearly every day” was eliminated in favor of relevant symptoms being present “during the same 2-week period.” Anhedonia, or decreased ability to take pleasure in normally enjoyable activities, was expanded to include loss of pleasurable affect or pleasure associated with social and other activities. In addition, “Markedly diminished interest in… activities” was thought to correspond too closely to apathy, a separate neuropsychiatric symptom seen in dementia, and was removed from the criteria. Finally, there were two additional criteria added: social isolation/withdrawal and irritability. The NIMH-dAD has been validated in a small cohort study³⁷ where it is seen to have higher sensitivity and specificity for depression than the DSM criteria.

While much research continues to use DSM criteria for depression in dementia, the NIMH-dAD criteria may be more clinically useful when differentiating depression from age-related physiologic and cognitive changes.

RATING SCALES

There are several rating scales used to evaluate dementia in depression. Each has its benefits and drawbacks. The Cornell Scale for Depression in Dementia (CSDD) was designed specifically for assessing depression in patients who have cognitive impairments that may hamper answering questions about depressive symptoms. It comprises 19 caregiver and clinician-rated items, a strength as individuals with dementia may not be able to accurately describe their own mood or symptoms. The Geriatric Depression Scale (GDS) is commonly used in the clinical setting. There are 15-item and 30-item versions and the former generally takes 5 to 7 minutes to fill out. The GDS focuses on cognitive and affective aspects of depression, avoiding somatic and sexual symptoms that may return false positive in an older population. The Hamilton Depression Scale’s (HDS) longer length and complexity renders it more useful for clinical studies than office use. It is interviewer-administered questionnaire better suited to monitor depressive symptoms and treatment efficacy, than to be used as a screening device (Table 5-2).¹³

COMORBIDITIES

Any evaluation of mood in dementia requires careful assessment of contributing medical comorbidities, medications with sedating or cognitively dulling side effects, inadequately addressed pain, and metabolic disturbances. Relevant tests include chemistry profile, complete blood count (Box 5-4), liver function tests, thyroid panel, vitamin B12 and folate levels. Additional workup to be considered includes a urinalysis, chest x-ray, electrocardiogram, and neuroimaging. In cases where a patient has episodic changes in mood or alertness, an electroencephalogram may be useful.

DIFFERENTIAL DIAGNOSIS

The reversible dementia of depression (formally termed “pseudodementia”) presents almost identically to a dementia caused by neurodegenerative processes. It is often diagnosed retrospectively, after successful treatment for mood symptoms results in resolution of cognitive deficits as well. Neuropsychological testing is considered the gold standard in distinguishing between the entities,³⁸ although “poor effort” on testing in the severely depressed may limit its usefulness. Apathy is a common finding in AD, and can present early in the course of the disease. Apathetic patients may appear disinterested in activities they used to enjoy, resist participation in social or family events, and remain inactive for extended periods of time. While apathetic patients have many of the same motivational deficits seen in depression, the syndromes are dissociable.^39,40 There are several scales designed to assess for apathy; these include, but are not limited to, the Starkstein Apathy Scale, the Lille Apathy Rating Scale and the Apathy Evaluation Scale. Another important differential diagnosis of dAD is pathological affective display.⁴¹ Individuals may present with sudden episodes of laughing or crying. The emotional reaction may be completely unrelated to the individual’s state of mind or the current situation. At other times, it may be congruent with the individual’s mood but out of proportion to the situation. It is important to note that while the crying may be a pathological display of affect, studies have found significantly higher levels of depression and anxiety in patients with pathological affect than in those without it.⁴² Thus, any individuals in whom pathological display of affect is suspected should be carefully screened for concomitant depressive symptoms (Box 5-5).

TREATMENT

There are compelling reasons to treat depression in patients with dementia as soon as possible. It has been demonstrated that depression is associated with earlier placement into a nursing home and a greater likelihood of being discharged earlier from an assisted living facility to a nursing home.^43,44 It is also associated with greater rates of physical aggression toward caregivers.⁴⁵ Finally, the caregivers for people with dementia and depression are at higher risk for depression themselves.⁴⁶ Promptly identifying and treating depressive symptoms results in better outcomes for both dementia patients and their caregivers.^47,48 Treatment may involve biological and psychosocial interventions. Box 5-6 summarizes treatment alternatives for depression in neurological illnesses.

PSYCHOPHARMACOLOGICAL TREATMENT

Antidepressants

Healthcare providers selecting a pharmacological intervention must take several factors into account, including medication’s mechanism of action, side effect profile, interactions, cost and dosing schedule. Serotonin reuptake inhibitors (SSRIs) are generally well tolerated, but may induce agitation, worsening apathy, tremor or sleep disturbances. In addition, it is particularly important to monitor interactions due to the heightened risk for serotonin syndrome, and EKG due to risk for QTc prolongation. SNRIs are an alternative with the potential added benefit of their noradrenergic action, although there are no controlled studies for its use in this population. Tricyclic antidepressants (TCA) carry cardiac risks, including conduction problems and potential heart block. They can also have anticholinergic side effects that interfere with cognition, worsen confusion or cause delirium. Other side effects include orthostatic hypotension, dizziness, dry mouth, constipation, sedation, and urinary retention.⁴¹ TCAs should be avoided in patients with closed angle glaucoma. If a TCA must be used, nortriptyline and desipramine have the least severe anticholinergic side effects and one can follow plasma levels to avoid toxicity. Monamine oxidase inhibitors (MAOI) have the benefit of little anticholinergic side effects. They are generally quite well tolerated, but require adherence to a restrictive diet, as ingesting tyramine-rich foods while on an MAOI can cause a hypertensive crisis. One way to circumvent this risk is by using the selegiline patch at its lowest dose of 6 mg/24 hours.¹³

Research studies have not provided clear direction about which medications confer greatest benefit to depressed individuals with dementia. A 2007 meta-analysis reviewed dAD treatment with TCAs and selective SSRIs. Overall, antidepressant treatment was superior to placebo.⁴⁸ However, a recent large community-based, multicenter, randomized/placebo-controlled controlled trial in patients with mild to moderate depression in AD showed no difference in patients treated with sertraline or mirtazapine as compared to placebo.⁴⁹ The Work Group on AD and other dementias (convened by the American Psychiatric Association) suggests SSRIs as the first-line treatment. This is mainly due to SSRIs being better tolerated than other antidepressant classes. There is no clear consensus on which SSRI works best in the dAD population, but paroxetine should be avoided for its relatively greater anticholinergic effects. If patients with dementia develop intolerable side effects or fail to achieve remission, one can consider substituting bupropion, an SNRI such as venlafaxine, or mirtazapine.⁵⁰

Antipsychotics

The use of antipsychotics in persons with dementia remains controversial. They do not have an indication for depression per se, but may be helpful in addressing comorbid agitation, aggression or psychotic symptoms. Conventional antipsychotics are generally avoided as older patients and those with neurodegenerative disorders may be especially vulnerable to side effects, such as tardive dyskinesias, parkinsonism, and increased risk of falling.⁵¹ Second-generation, or atypical, antipsychotics are better tolerated but several well-documented studies have shown that they may carry an increased risk of adverse cerebrovascular events and a higher rate of all-cause death in individuals with dementia.⁵² However, as most seasoned geriatric psychiatrists and behavioral neurologists can attest, these concerns must be balanced with the need to manage the neuropsychiatric symptoms of dementia that may be dangerous to the patient or their caregivers. In 2008, the American College of Neuropsychopharmacology (ACNP) published a white paper reviewing available evidence on antipsychotic drug use in the elderly with dementia.⁵³ They provided several clinical recommendations, which included: ruling out medical etiologies of BPSD (e.g., delirium or pain), utilizing nonpharmacologic interventions, and involving caregivers and family members in informed decision making about when to use antipsychotics and the potential risks. If antipsychotic use is unavoidable, the authors recommended identifying target symptoms, monitoring for efficacy and side effects frequently, and discontinuing the medication if target symptoms did not decrease or resolve.

In regard to medication choice, the authors felt that there was not enough evidence to suggest one atypical over another. Whatever agent was used, they advocated for the lowest required doses for the shortest time possible.

Other Psychopharmacotherapy

Studies that looked at the efficacy of cholinesterase inhibitors for neuropsychiatric symptoms in dementia have demonstrated improvement in various mood symptoms (e.g., anxiety, apathy) but not depression.^54–56 Similarly, randomized control trials of anticonvulsant mood stabilizers have provided mixed results at best.⁵⁷ Stimulants have also failed to show conclusive effects on depression of dementia: while one study found improvement in depression scores in a small group of AD patients treated with methylphenidate,⁵⁸ another suggests that this may be due to a reduction in negative symptoms rather than true remission of depressive symptoms.⁵⁹

BRAIN STIMULATION THERAPIES

ECT can be considered to address depressive symptoms in dementia.⁶⁰ In the elderly without dementia, it can be quite useful in treating depression with psychotic features. However, dementia patients are at heightened risk for adverse cognitive effects of ECT, especially those at later stages of a neurodegenerative process or with a heavy cerebrovascular burden. Often times, dementia patients have medical comorbidities rendering them vulnerable to anesthetic agents. Thus, ECT should be reserved for medication refractory cases, and the patient and/or family must be intensively counseled on the risks and benefits of ECT before providing consent.

There are a limited number of studies looking at other brain stimulation therapies’ effects on depression in dementia. There is no consensus as to whether deep brain stimulation (DBS) has any effect on neuropsychiatric symptoms of AD, although its effects in PD are discussed later in this chapter. Vagus nerve stimulation (VNS) appears to be well tolerated in the Alzheimer population but has yet to demonstrate robust effects on mood symptoms.^61,62 Transcranial magnetic stimulation (TMS) may have a mild antidepressant effect in depressed patients with AD^63,64 and dementia with Lewy bodies.⁶⁵

PSYCHOSOCIAL TREATMENT

The psychosocial treatment of depression in dementia is discussed in a separate section below.

SUMMARY

In summary, first-line pharmacologic treatment of depression in dementia should be an antidepressant. Among the various classes, SSRIs offer the best risk/benefit ratio, given their relatively benign side effect profile and lack of diet restrictions. SNRIs can be a useful alternative. If depressive symptoms are refractory, one can consider TCAs or MAOIs with careful patient and caregiver education. ECT remains an option for severe depression when medication truly cannot be tolerated or is ineffective. While newer brain stimulation therapies such as VNS and TMS appear to be well tolerated, they have yet to build a robust body of evidence to support their use in treatment depression in dementia.

PSYCHOSOCIAL TREATMENT OF DEPRESSION IN DEMENTIA AND OTHER NEUROLOGICAL DISEASES

While the main focus of this section is psychosocial treatment of depression in Dementia, many of the general principles discussed here can be applied to depression in other neurological conditions; this includes depression in the context of other neurodegenerative illnesses such as PD, and depression related to brain vascular illness. Often it is family members and not the patient him- or herself who request help with dementia diagnosis and treatment. It is important to help families understand that anosognosia represents impairment in the cognitive processes that support insight; in other words, “denial” is not purely a psychological defense.⁶⁶ While there is variability in the response to dementia diagnosis, an overarching theme is the tension between preserving key aspects of identity and accommodating change; for people with dementia and their caregivers, dynamics of autonomy and independence are salient with conflicts often emerging around safety.⁶⁷

Consensus has emerged in favor of “telling the truth in dementia” and current guidelines warn against providers’ fear of inflicting harm and caregiver-expressed concerns about impact of diagnosis on patients; instead, the recommendation is for an early dialogue with the patient to elicit his or her preferences for information on diagnosis and prognosis and involvement in treatment decision-making.⁶⁸ As preferences may change over time it is important to continue this conversation longitudinally, timing and tailoring the referrals according to the clinical needs and the range of resources available.

The majority of people with dementia live in the community with family caregivers who themselves are at risk for psychiatric morbidity (see Box 5-7).⁶⁹ Rates of depression in dementia caregivers range from 23% to 85% in developed countries.⁶⁹ The burden of care is highest in the presence of BPSD and neuropsychiatric symptoms, rather than cognitive impairment and needs for direct physical care, are more likely to prompt institutionalization (see Text box 5-7).^70–72

While there is evidence that support and counseling for caregivers reduces depression⁷³ the most effective interventions add cognitive restructuring techniques⁷⁴ and individually tailored behavioral management training.^71,75

Given risk of MCI advancing to dementia, MCI caregivers are ideal targets for skill- and resource-building interventions.⁷⁶ Spouse caregivers of people with early-onset Alzheimer’s and frontotemporal degeneration are hard-hit as they are often raising young children and suffer financial losses during wage-earning years⁷⁰; they are in need of more specialized information and support which may best be delivered by telephone and internet in remote areas where more specialized services are not available.⁷⁷ Technology-based formats may also better reach caregivers in cultural minorities who have lower rates of resource use.⁷⁸

The loss of language, memory, and insight inherent to dementia may be obstacles to the use of traditional psychotherapy. The number of randomized controlled trials of psychotherapy for depression in MCI and dementia is limited and the sample size in existing studies is small; however, research shows that psychotherapy can be adapted and is acceptable to people with MCI and early dementia.^79–82 There is some evidence that long-term interpersonal psychotherapy protects against depression recurrent in older people with mild to moderate cognitive impairment.⁸⁰ Even when adapted to include caregiver, short-term psychodynamic psychotherapy does not seem to produce measurable benefit.⁸³ More research is needed to determine which specific cognitive abilities predict or preclude psychotherapy.

Problem-solving therapy (PST), an approach of inculcating skills to cope with everyday problems, has been adapted to the treatment of depression in elderly people with dementia and executive dysfunction, and is more effective that more emotionally oriented supportive psychotherapy in treating depression and reducing disability.^84,85 It is important to note, however, that supportive therapy is a highly flexible approach that includes aspects of problem solving.⁸⁶ Problem-solving and behavioral therapy have been adapted for intervention with caregivers of people with dementia. The caregiver is trained to initiate pleasant events, distract from negative thoughts, and alter the environment/routine. In two randomized controlled trials, problem-focused therapy with the patient and caregiver in the home was more effective than supportive therapy in reducing symptoms of depression in the person with dementia.^87,88

The Alzheimer’s association offers community-based support and education programs for patients and caregivers, especially in the vulnerable period after initial diagnosis.⁸⁹ Participants meet with other newly diagnosed individuals in their community to share experiences and concerns, learn more about the disease, reduce feelings of isolation, and receive assistance in coping with lifestyle changes and developing long-term care plans. While some researchers caution that providing too much information too soon can increase depression for newcomers to groups,⁹⁰ a waitlist-controlled clinical trial of an Alzheimer’s Association Early Stage Memory Loss Support Group found significant reduction in depression symptoms for those enrolled in the group.⁹¹ A large-scale randomized controlled Danish trial of 12-month structured psychosocial intervention of counseling, education, and support to people with Alzheimer’s and their caregivers found only a small positive effect on mood which was not sustained 24 months after the intervention ended.^92,93 A similar Norwegian study found no effect.⁹⁴ The lack of findings may be attributed to a floor effect as patients had minimal depression symptoms at the outset.^92,93 These studies suggest that it is important to assess need and adjust level of intervention accordingly; doing so requires regular follow-up and the ability of providers to “prescribe interventions”⁹⁵

Although mood is not the primary target of cognitive rehabilitation, training, and remediation, these interventions have secondary benefits likely mediated by increasing experiences of competence and decreasing experiences of frustration in everyday life.^96,97 Incorporating cognitive behavioral interventions and motivational interviewing techniques may improve engagement and outcomes.⁹⁸ A waitlist-controlled study combining activity planning, assertiveness training, relaxation techniques, stress management, motor exercise, use of memory aids, and motor exercise in group format for people with MCI showed improvement in mood.⁹⁹ A randomized control trial of a cognitive-motor intervention consisting of reality orientation, ADL training, and psychosocial support produced mood and cognitive benefits in a people with Alzheimer’s compared with patients treated with psychosocial support alone.¹⁰⁰ Some of the highest-quality studies producing the most significant results combine different categories of intervention; this makes it difficult to determine the active ingredient but also suggests that a combination of interventions produces important synergies. Examples include a randomized controlled trial of 153 community dwelling people with AD which combined a home-based exercise program with caregiver training in behavioral management techniques and found significantly reduced rates of depression and delayed institutionalization.¹⁰¹ Psychosocial interventions can also be used in the treatment of depression and apathy in more advanced dementia and long-term care settings; these include multisensory stimulation, a variety of activities (music, dance, exercise), therapeutic conversation/reminiscence, pet therapy, cognitive stimulation (as distinct from rehabilitation)/reality orientation, and models of care. Individualized interventions that take into account characteristics and interests of the person with dementia and provide opportunities for meaningful activity are most effective in reducing symptoms of depression and apathy.^102–107 Music may be most effective in more advanced dementia.^104,108

SUMMARY

There are various challenges to interpreting the robust meta-analytic literature on psychosocial interventions.^{72,75,79,81,109,110} Researchers use different terminology to label interventions, making it difficult to directly compare individual referenced studies. Inclusion criteria vary among the reviews, especially for control conditions, raising concern about the placebo effect of receiving stimulation and human contact regardless of the effectiveness of any specific method.¹¹¹ Conversely, a finding of lack of effectiveness for a specific intervention may represent an artifact of difficulty validating an intervention.¹⁰⁶

Despite equivocation about the effectiveness of specific interventions and various over-arching approaches, comparison of the meta-analytic literature yields strong support for interventions that are individually tailored, use multiple treatment components, include the caregiver, and include follow-up.^72,75,111 There is particularly strong evidence for problem-solving approaches and modified cognitive behavioral therapy (CBT) intervening directly with a patient with MCI or early dementia or by training the family caregiver to deliver the interventions.^109,110 In line with the correlation between neuropsychiatric symptoms (see Text box 5-8) and caregiver stress, improving caregiver skills to manage behaviors also benefits caregiver mood.

Cerebrovascular disease is a significant contributor to late-life depression. The terminology in this section follows current usage in distinguishing poststroke depression from “vascular depression,” a term which in the past has been used to refer to depression related to both discrete infarcts and to the accumulation of subcortical microvascular ischemic changes. Poststroke depression is diagnosed based on the temporal relationship between a clinically apparent stroke and the onset of depression. Vascular depression is an evolving concept that arose from the observation of the correlation of depression and subcortical microvascular disease, which has become more widely appreciated in the age of increasingly available MRI. Unlike idiopathic MDD, poststroke and vascular depression primarily have onset late in life. They have modifiable risk factors, which coincide with the risk factors for cerebrovascular disease itself. Anticipation, recognition, and treatment of depression associated with cerebrovascular disease can reduce its morbidity, and improve outcomes related to vascular disease itself (see Text box 5-9 for a review of Anxiety and Cerebrovascular disease).¹¹²

EPIDEMIOLOGY

Stroke is the fourth leading cause of death in the United States, and it is estimated that 6.8 million Americans over the age of 20 have suffered a stroke, with an annual incidence of approximately 795,000¹¹³ Among the approximately 75% who survive a stroke, one-third will experience poststroke depression, with the greatest risk in the first year after the stroke.¹¹⁴ See Figure 5-1 for a comparison of the prevalence of depression in different neurological illnesses and Box 5-1 for risk factors for depression in neurological disease. The factors that increase the risk of depression after stroke include: personal or family history of depression, degree of subcortical atrophy prior to the stroke, and degree of physical or cognitive impairment after the stroke. Confounding the last factor, however, is the observation that the presence of poststroke depression worsens outcomes and increases the degree of residual disability after stroke.¹¹⁵

PATHOPHYSIOLOGY

Several factors may mediate the development of depression in neurological illnesses (Box 5-2). The experience of stroke as a sudden loss of physical, verbal, or cognitive function, and the immediate onset of significant disability is a traumatic experience, and poststroke depression shares some of its origins in the reactive nature of depression that emerges in the wake of any significant medical diagnosis or illness. The fact that stroke is itself a brain injury, however, suggests that depression may arise from the infarction itself. This hypothesis has been borne out in the observation that the risk of poststroke depression is related to the location of the lesion. The ability of lesion location to predict poststroke depression remains debatable, but evidence suggests that infarcts in the left frontal lobe and basal ganglia are more likely to precipitate depression. Poststroke depression may correlate with the proximity of the infarct to the anterior pole of the left frontal lobe¹¹⁶ though this is not consistently observed, and may be more significant in the early phases of stroke recovery (2–6 months after the event).¹¹⁷

A depletion of monoamine neurotransmitters may also be involved in the development of poststroke depression. Multiple rodent models of stroke have shown ipsilateral depletion of serotonin, norepinephrine, and dopamine after stroke, and reduction of monoamine metabolites has been demonstrated in the CSF of human patients with poststroke depression. PET imaging of 5-HT2 receptors shows evidence of greater upregulation of serotonin receptors in the right hemisphere compared to the left in patients with poststroke depression. This difference in receptor expression after stroke may represent hemispheric differences in response to injury, and correlate with the lateralization of depression risk after stroke, as well as support the hypothesis of monoamine depletion leading to poststroke depression.¹¹⁸

CLINICAL PRESENTATION

Symptoms of poststroke depression do not significantly differ from those of idiopathic major depressive disorder. Because of overlapping symptoms of stroke and depression (e.g., changes in energy, sleep, appetite, libido, and cognition), several studies have examined the validity of DSM criteria for major depression in the setting of stroke. Adjusting diagnostic criteria, however, to account for the origin of neurovegetative symptoms (i.e., attempts to exclude symptoms judged to be direct sequelae of the stroke) does not improve the sensitivity or specificity of DSM criteria for major depressive disorder in the setting of stroke.¹¹⁹

DSM criteria for MDD are sensitive and specific for poststroke depression as well. The symptoms of poststroke depression do not differentiate it from other forms of late-life depression. The particular presentation of individual cases likely depends on the lesion location and the extent of disability, combined with the individual’s reaction to the stroke due to the new and potentially traumatic onset of functional impairment.

COURSE AND NATURAL HISTORY

The course of poststroke depression has been examined in several longitudinal studies, but the conclusions have been inconsistent and the degree of treatment in most case series has not been clearly delineated. There is some consensus, however, that depressive symptoms that emerge rapidly, within hours to days of stroke, tend to peak within 3 to 6 months of onset and approximately 50% experience remission by 1 year. Patients who experience onset of depression 2 months after stroke or later, however, typically have a more protracted course, and up to 50% remain depressed 2 years after their stroke. One confounding factor is the degree of physical disability, which correlates with risk of depression; patients with greater disability tend to be overrepresented in the hospital and rehab settings where patients have been recruited for many of these studies.¹²⁰

ASSESSMENT AND DIFFERENTIAL DIAGNOSIS

Lack of awareness of the high prevalence of poststroke depression, misattribution of depressive symptoms to physical consequences of the stroke, and perception that the patient’s distress may be an “appropriate” reaction to stroke all likely lead to the underdiagnosis of poststroke depression. Recognition of this entity, however, is important, because appropriate treatment of depression can alleviate the suffering of the depressed patient as well as improve rehabilitation outcomes. As discussed, DSM criteria for MDD are sensitive and specific for poststroke depression. The defining characteristic of this entity is onset after a clinically apparent stroke. Other symptoms can emerge after stroke, however, that can complicate or confound the diagnosis of depression. The most common poststroke symptom that mimics depression is apathy. By definition, apathy is a reduction of motivation not attributable to emotional distress, cognitive impairment, or level of consciousness.¹²¹ Apathy is also a common feature of depression, and is included on the Hamilton Rating Scale for Depression. However, several studies and reviews demonstrate that apathy and depression are not always correlated and can be differentiated.¹²² A case series¹²³ compared apathy and depression levels across stroke, AD, and idiopathic major depression. While the relationship between apathy and depression varied among the groups, patients with major depression or left-hemisphere stroke tended to higher depression scores and lower apathy scores. Patients with right hemisphere stroke had high levels of both apathy and depression, although these symptoms did not correlate with each other in this group. A longitudinal study of patients with poststroke depression found that within 3 months of a stroke, levels of apathy and depression did not correlate, but that a correlation emerged over time and was significant at 1 year. Both apathy and depression were predicted by the presence of dementia, but depression was independently predicted by psychosocial factors, such as not living with a family member. Although the correlation increased, there were a significant number of patients at 1 year who demonstrated apathy or depression, but not both.¹²⁴

Although apathy likely shares with depression an origin in the disruption of frontal subcortical networks, distinguishing the two phenomena has treatment implications: failure to distinguish them can lead to treatment failure. Primarily studied in dementia, apathy has not been found to be responsive to antidepressants,¹²⁵ whereas poststroke depression has (discussed below). When apathy occurs after stroke, careful assessment for other symptoms of depression is required to differentiate the two syndromes. If apathy coexists with depression, the treatment of both syndromes may be indicated.

TREATMENT

Treatment of poststroke depression is important, as degree of depressive symptoms negatively correlates with rehabilitation potential¹¹² and poststroke outcomes. See Box 5-6 for a summary of treatment of depression in neurological disorders.

PSYCHOPHARMACOLOGICAL TREATMENT

Numerous studies comparing antidepressants against alternate agents or placebo have demonstrated widely varying efficacy in antidepressant treatment for poststroke depression. The range of study types, enrollment criteria, and means of assessment make it difficult to draw firm conclusions, but several studies have demonstrated efficacy of antidepressant agents in this population. The most commonly studied agents are SSRI and TCAs. In general, antidepressants from either class outperform placebo, although placebo response is high, which is typical for antidepressant trials. Nortriptyline has been shown to significantly outperform both fluoxetine and placebo.¹²⁰ Use of tricyclic agents, while possibly more effective than SSRIs or SNRIs, is limited by their adverse effects. Patients with poststroke depression tend to be elderly and are more likely to have vascular disease. Anticholinergic effects and risk of cardiac arrhythmias may lead most clinicians to reject these agents in favor of SSRI antidepressants.

Several case series and chart reviews have demonstrated the safety and tolerability of psychostimulants in poststroke depression, but have not demonstrated efficacy. Nonetheless, the potential of stimulants to increase energy and physical activity level has led to their empiric use, particularly in the rehab environment.

BRAIN STIMULATION THERAPIES

Two retrospective chart reviews have demonstrated the safety and efficacy of ECT for poststroke depression, with some patients receiving treatment within 1 month of stroke.¹²⁶ Repetitive transcranial magnetic stimulation (rTMS) has growing evidence for use in recovery of motor and cognitive stroke symptoms, but no published evidence to date addresses its use for poststroke depression.

PSYCHOSOCIAL TREATMENT

Several of the principles discussed above, regarding the psychosocial treatment of depression in dementia, also apply to poststroke depression. Poststroke patients also struggle with issues of loss of function and many times loss of independence. They may depend from caregivers and the inclusion of caregivers in treatment may increase effectiveness of the interventions. On the other hand, in the context of a static brain lesion, the potential for rehabilitation changes the prognosis and the dynamics of the relationship with caregivers. Different approaches including cognitive behavioral therapy, mindfulness-based interventions, and acceptance and commitment therapy have been used. Unfortunately the evidence of benefit of psychotherapy in this population has been limited.¹²⁷

SUMMARY

In addition to significant physical disability stroke can also lead to depression, which can be an independent cause of disability and also exacerbate the residual physical symptoms of a stroke. Particularly in frontal lobe strokes, the lesion itself may injure regions of the brain involved in emotional regulation, leading directly to depression. Poststroke depression has been associated more with infarcts of the left frontal lobe than with other brain regions, but depression can emerge after an infarct in any location. The sudden onset of significant disability and loss of independence after a stroke can trigger a depressive episode. Patients with personal and family histories of depression are at higher risk. Recognition and treatment of depression in the wake of a stroke, particularly in the acute phase, can improve rehabilitation participation and outcomes. Although TCAs have more evidence for efficacy in poststroke depression their side effects may limit their use, particularly in elderly patients. SSRIs have demonstrated efficacy as well, and ECT has been used successfully used in severe and refractory depression.

Although cases of depressed elderly patients with atherosclerosis were described prior, “vascular depression” was first hypothesized as a distinct syndrome when Alexopoulos et al. described an association of late-life depression with impaired executive function and white matter hyperintensities on T2-weighted MRI.¹²⁸ This suggested areas for further investigation including involvement of frontal subcortical circuits in idiopathic depression, and also evolved into a recognizable subtype of depression. However, there is not yet a consensus for criteria defining vascular depression, which complicates attempts to investigate the syndrome. Some investigators prefer the term “subcortical ischemic depression,” and others refer to “depression executive dysfunction,” which describes the syndrome symptomatically allowing for other causes. Despite terminology differences, most investigators agree that the triad of depression onset after sixth decade, white matter hyperintensities found on MRI, and executive dysfunction comprises the core features of vascular depression.

EPIDEMIOLOGY

Estimating the prevalence of vascular depression is complicated by the general under-recognition of late-life depression and the requirement of imaging and cognitive assessment for diagnosis. In a large cross-sectional study vascular depression prevalence is estimated at 3.4% of Americans aged 50 and older, approximately 2.64 million people.¹²⁹ However, the criteria in this study do not include imaging findings or measures of cognitive impairment, and include poststroke depression. Other salient findings include evidence of late-life depression under-treatment and increased morbidity of cerebrovascular disease-associated depression in which only 40% of the depressed respondents reported current treatment, and only 10% reported treatment considered adequate to guidelines. Respondents with cerebrovascular disease or risk factors reported significantly higher functional limitations across several domains, including cognitive, mobility, social, and overall role impairments.

PATHOPHYSIOLOGY

See Boxes 5-1 and 5-2 for a summary of factors relevant to the development of vascular depression. Studies on the mechanism of vascular depression have centered on the white matter hyperintensities which represent microvascular disease. The amount of white matter hyperintensities correlates with age, regardless of the presence of depression. There is a correlation between the white matter disease burden in the frontal lobes and the incidence of depression. Periventricular hyperintensities appear equally prevalent in depressed and nondepressed subjects. Deep white matter hyperintensities, however, have consistently been found to be more prevalent in depressed subjects, and with late-onset depression in particular.¹³⁰ These deep white matter lesions are thought to be disruptive to frontal-subcortical and to a lesser extent temporal lobe function in vascular depression.

The correlation of executive dysfunction with white matter disease and depression suggests disruption of dorsolateral prefrontal striatal circuits. Postmortem tissue analysis demonstrates that subcortical ischemia preferentially affects the dorsolateral prefrontal cortex in patients with late-life depression¹³¹ and that white matter hyperintensities in vascular depression are ischemic, rather than inflammatory.¹³²

CLINICAL PRESENTATION

Vascular depression differs from other forms of depression in several ways, although there is significant overlap.¹³³ Anhedonia as a major symptom is more common in vascular depression than subjectively depressed or low mood (Box 5-3). Vascular depression shares characteristics with frontal lobe syndromes, particularly those arising from dysfunction of medial or dorsolateral prefrontal cortices. Executive dysfunction, commonly seen in dysfunction of the dorsolateral prefrontal cortex or its connections, is a defining characteristic of vascular depression. Family history of mood disorder is less common in vascular depression than in idiopathic depression.

COURSE AND NATURAL HISTORY

The clinical course of vascular depression is similar to that of refractory major depression in that it generally becomes a chronic condition and tends not to respond to antidepressant treatment. Progression of white matter disease is a risk factor for the onset of vascular depression¹³⁴ but it is not known whether progression of subcortical vascular disease in an individual patient correlates with worsened depression. Vascular depression is often comorbid with dementia, as microvascular disease causes subcortical vascular depression and also exacerbates dementia from other etiologies, such as AD.

ASSESSMENT AND DIFFERENTIAL DIAGNOSIS

The presence of executive dysfunction in vascular depression may lead to some confusion over whether a given patient’s symptoms represent depression or dementia. The syndrome of “reversible dementia of depression,” also known as “pseudodementia,” has long been described, with deficits not only of memory but also of executive dysfunction (Box 5-5). Concern over this syndrome has been that clinicians often overlook depressive symptoms in the presence of cognitive impairment in elderly patients. Awareness of the entity of vascular depression, however, introduces a third consideration in addition to the possible diagnoses of idiopathic depression and dementia due to neurodegenerative process. Vascular depression comprises symptoms of both depression and executive dysfunction and shares features with depression and dementia of other etiologies. Diagnosis of vascular depression, however, implies the presence of microvascular disease, demonstrated by subcortical white matter hyperintensities on T2-weighted MRI. Brain imaging in this setting may help distinguish among idiopathic depression and vascular depression in elderly patients or patients with significant vascular risk factors.

The syndrome of apathy can also present a diagnostic challenge. Apathy and executive dysfunction may present as features of vascular depression, but also can signify the emergence of a neurodegenerative disorder or the sequelae of a stroke or other structural lesion. Differentiation of apathy from depression has been previously covered in the discussion of poststroke depression, and similar considerations apply with vascular depression.

TREATMENT

PSYCHOPHARMACOLOGICAL TREATMENT

Given its likely distinct psychopathology, it seems intuitive that the treatment profile of vascular depression would differ from idiopathic depression. Several studies have demonstrated variable response to SSRIs. Response to these medications seems to inversely correlate with the progression of white matter hyperintensities¹³⁴ and degree of neuropsychological impairment. In a nonrandomized trial 33% of patients achieved remission over 12 weeks with SSRI treatment. The likelihood of remission, assessed by MADRS score, diminished with increasing deficits in overall executive function, language processing, episodic memory, and processing speed.¹³⁵ A placebo-controlled trial of sertraline in elderly patients with depression (but not specifically with vascular depression) found that sertraline performed worse than placebo in patients with impaired executive function, as measured by response inhibition.¹³⁶ This suggests that patients with executive dysfunction and depression, including vascular depression, may be subject to the side effects of antidepressants but not their benefits.

BRAIN STIMULATION THERAPIES

A study comparing rTMS of the prefrontal cortex to sham rTMS found a response rate of 39% and a remission rate of 27%,¹³⁷ which are comparable to rates seen with antidepressants. Unlike antidepressants, however, response to treatment did not correlate with degree of cognitive impairment or executive dysfunction. There was a decreasing response to rTMS with increasing age and also with decreased volume of frontal gray matter. There is limited evidence regarding use of electroconvulsive therapy (ECT) in vascular depression, but one small series and several case reports suggest it is effective and generally well tolerated, although perhaps with increased risk of delirium compared to ECT in idiopathic depression.¹³⁸

PSYCHOSOCIAL TREATMENT

Given its usual age of onset, psychotherapy with patients with vascular depression shares characteristics with psychotherapy with the ill, elderly patient; please refer to Chapter 17 for discussion of this topic. An important role of psychotherapy with this population is to support the work patients need to do in sustaining positive health behaviors, to help decrease their vascular risk factors. Several of the aspects of psychotherapy in the context of dementia, discussed above, apply to this population as well; importantly, the inclusion of caregivers in the plan of care may augment treatment results.

SUMMARY

Vascular depression is an increasingly recognized etiology of late-onset depression, often presenting with cognitive symptoms characteristic of subcortical dementia. The correlation of depression severity and cognitive dysfunction with the extent of subcortical microvascular disease seen on MRI suggests that disruption of subcortical networks is the common cause of both the depression and cognitive symptoms. Response to antidepressants appears to be poor in vascular depression. Attention, therefore, should turn to prevention. Although not proven, prevention of cerebral microvascular disease through blood pressure control, management of hyperlipidemia, and other vascular risk factors, may offer protection against vascular depression in addition to its known other cerebrovascular and cardiovascular benefits.

EPIDEMIOLOGY

PD is a neurodegenerative illness resulting from deterioration and loss of dopamine-producing neurons in the substantia nigra of the midbrain. This results in decreased dopaminergic input to the basal ganglia, producing the characteristic motor symptoms of rigidity, bradykinesia, tremor, and postural instability. It is the second most common neurodegenerative disease after AD. Estimated prevalence in developed countries is 0.3% of the population and 1% of those 60 and older. Prevalence increases with age up to about 4% in the oldest cohorts,¹³⁹ and nearly 1 million Americans have PD.

In addition to motor manifestations, psychiatric symptoms are common, with depression being the most frequent (Fig. 5-1). Estimates of depression rates in PD vary widely, but it is estimated that 17% of patients with PD meet DSM-5 criteria for MDD, with an additional 35% meeting criteria for dysthymic or subsyndromal depression.¹⁴⁰

Rates of depression in patients who are ultimately diagnosed with PD are about double that of the general population.^141,142 Mood symptoms can precede clinical motor parkinsonism for up to 20 years, but the peak is 3 to 6 years prior to onset of motor symptoms and PD diagnosis.¹⁴³ It should be noted that most reports of depression onset rely in part upon patients’ recall and thus are subject to some degree of reporting bias. However, an association between depression and PD is upheld by a retrospective cohort study of patients with depression that¹⁴⁴ found a hazard ratio for developing PD of 3.13 among patients diagnosed with depression using International Classification of Primary Care (ICPC) criteria.

PATHOPHYSIOLOGY

It is likely that some cases of depression associated with PD are reactions to the diagnosis and long-term prognosis of progressive disability (see Box 5-2). However, prevalence of depression in PD is as much as twice that seen in other conditions that produce equivalent disability¹⁴⁵ suggesting that the pathophysiology underlying PD contributes to the development of depressive symptoms.

There is a correlation between depressive symptoms and certain motor symptoms, which supports the idea of a common pathophysiologic pathway. For example, there is a higher prevalence of depression in patients with predominantly akinetic-rigid motor symptoms versus tremor-rigid-bradykinetic symptoms.¹⁴⁶

The primary pathophysiological process in PD is regression and loss of dopamine-producing neurons in the substantia nigra. The downstream effects of this include motor symptoms due to loss of dopaminergic input to the basal ganglia. The basal ganglia, in particular the ventral striatum, also have significant function in emotional processing, as part of the prefrontal-striatal-thalamic circuits. PET imaging of dopamine and norepinephrine transport receptors in patients with PD and depression¹⁴⁷ shows that depression symptom severity, measured by Beck Depression Inventory (BDI), correlates with decreased dopamine and norepinephrine receptor binding in the ventral striatum. Patients with PD and depression have reduced binding in the locus coeruleus, mediodorsal and inferior thalamus, left ventral striatum, and right amygdala when compared to patients with PD and without depression. The latter group shows less binding in these areas compared to healthy controls. This suggests that depression in PD may be a consequence of a subset of the pathological processes of PD itself. It is not clear whether this is related to an underlying predisposition or vulnerability, or from idiosyncratic disease progression.

Structural and functional imaging studies further highlight the involvement of the mediodorsal, or “limbic” thalamus. In an fMRI and voxel-based morphometry study comparing patients with PD with or without depression, there is increased volume in bilateral mediodorsal thalami and reduced activation in the left mediodorsal thalamus and in the left medial prefrontal cortex in the depressed patients.¹⁴⁸

CLINICAL PRESENTATION

Depression may emerge before diagnosis or in the early or late stages after diagnosis of PD. Somatic symptoms of depression may be masked or confounded by parkinsonian motor symptoms, such as bradykinesia and hypomimia (Box 5-3). Many patients with PD who do not have depression experience early waking and decreased energy, and psychomotor retardation is nearly universal. Other symptoms, however, such as decreased appetite and libido, increased sleep latency and overnight waking, and nonsomatic depression symptoms such as low mood, anhedonia, feelings of guilt, and preoccupation with death are generally attributable to depression.¹⁴⁹ PD patients on levodopa may experience dysphoria and/or anxiety, combined with their motor fluctuations, as part of the “on/off” cycle: In most cases the effect of levodopa gradually wears off a few hours after each dose. During this “off” period the patient experiences exacerbation of the PD motor symptoms, and may also become more depressed and anxious. After the following levodopa dose, in the “on” period, the patient’s mobility increases—possibly with added dyskinesia—and the mood improves.

COURSE AND NATURAL HISTORY

Evolution of depressive symptoms in PD tends to mirror the motor symptoms. A longitudinal study of nonmotor symptoms¹⁵⁰ shows that depressive symptoms decrease over the first 2 years after PD diagnosis as dopaminergic therapy is initiated and titrated. As neurodegeneration progresses, however, depressive symptoms return and become more refractory, as do the motor and autonomic symptoms.

Depression in PD can exacerbate the disability produced by motor symptoms, and itself can be considered a marker of disease severity. Depression is independently correlated with degree of disability in PD, and in a cross-sectional study¹⁵¹ nonmotor symptoms (primarily depression and cognitive impairment) account for 37% to 54% of the total variance in disability among patients.

Patients recognize the morbidity associated with depression in PD. In a patient survey,¹⁵² patients with early (<6 years) PD rate “mood” as the sixth most bothersome symptom. Patients with more advanced illness (>6 years) however, rate “mood” as the second most bothersome symptom, behind medication-related symptom fluctuations.

ASSESSMENT AND DIFFERENTIAL DIAGNOSIS

There are no specific criteria for diagnosis of depression in PD, and DSM criteria are generally used. Marsh¹⁴⁹ has suggested modifying DSM criteria somewhat to be more inclusive of the range of depressive symptoms found in PD. She recommends disregarding the “etiologic” criteria of DSM and considering symptoms based on observation rather than presumed etiology. Another recommendation is to carefully assess anhedonia to distinguish it from apathy, which may appear in patients with PD who are not depressed. Last, diagnosis of minor depression or dysthymia in PD is encouraged, as these “subsyndromal” depressive symptoms can cause significant distress and functional impairment in PD, without rising to the level of major depression.

Depression scales may be useful in PD to monitor the progression of symptoms or to screen for unreported or difficult-to-detect depression. The BDI¹⁵³ and the Hamilton and Montgomery–Asberg rating scales¹⁵⁴ are valid instruments for diagnosing and assessing severity of depression in PD, although the cutoffs for clinical significance tend to slightly overdiagnose depression because of the confounding effect of motor symptoms.

Because of its inherent morbidity and association with greater functional impairment, diagnosis of depression in PD is as important as it can be difficult. The overlap of motor symptoms with depression symptoms can easily lead to depression being overlooked. In one prospective study, diagnostic accuracy of the treating neurologist was 35%.¹⁵⁵ Routine screening of PD patients for depression and other nonmotor symptoms should be commonplace and is effectively done using self-report checklists, such as the BDI.

TREATMENT

PSYCHOPHARMACOLOGIC TREATMENT

Treatment of depression in PD has evolved in recent years (Box 5-6). Even in the absence of controlled trials, clinicians have used antidepressants to treat the observed symptoms of depression in patients with PD, with some anecdotally encouraging results. As most psychiatrists switched from TCAs to the better-tolerated SSRIs, most neurologists and neuropsychiatrists also switched to SSRIs for their patients with PD who were depressed. There have been limited controlled trials of antidepressants in this population and results have been inconsistent. A study in 2009 that compared nortriptyline with paroxetine for depression in PD shows significant advantage of nortriptyline,¹⁵⁶ although this study had several limitations including its brief duration, relatively small size, and a high dropout rate.¹⁵⁷

There is some concern over the potential exacerbation of motor symptoms by SSRIs, however the evidence remains inconclusive.^158,159 A randomized, double-blind, placebo-controlled study demonstrates that both the SNRI venlafaxine and the SSRI paroxetine may improve depression in subjects with PD.¹⁶⁰ The use of another SNRI medication, duloxetine, has benefit in a small series.¹⁶¹ The positive effects of SNRIs are not surprising, as the increase in both serotonergic and noradrenergic tone produced by these medications is similar to the actions of TCAs. Bupropion, an antidepressant which inhibits reuptake of norepinephrine and dopamine (NDRI), should theoretically be of benefit in PD, given its dopaminergic mechanism of action. It has been reported effective anecdotally and in a single published case report, and recommended for further study because of its low likelihood of worsening, and potential for ameliorating, motor symptoms.^162,163

Treatment of PD itself, with dopaminergic medications, can also treat concomitant depression. MAO inhibitors, while they have fallen out of fashion as antidepressants, continue to be used to treat PD, typically providing benefits in reducing “on/off” motor fluctuations. Most commonly used are the selective MAO-B inhibitors selegiline and rasagiline. One trial comparing two doses of rasagiline in 22 patients with PD and depression shows significant improvement in motor symptoms, without significant differences between patients receiving 1 or 2 mg/d. All patients also showed improvement in depression symptoms, however, the patients receiving the higher dose show significantly greater improvement in depression as measured by the Hamilton Depression Rating Scale.¹⁶⁴ This suggests that the antidepressant benefits of rasagiline are independent of its motor benefits, and that this medication may be particularly useful in patients with PD who are depressed. Dopamine agonists such as pramipexole, commonly used to treat PD’s motor symptoms, may also benefit PD depression. The evidence from controlled studies is still insufficient to recommend these agents as a first line of treatment.

BRAIN STIMULATION THERAPIES

While there has not been a randomized controlled trial of ECT for depression in PD, there are numerous noncontrolled studies reporting efficacy and tolerability. ECT should certainly be considered as a treatment for medication-refractory or severe depression in patients with PD. rTMS has shown promising results in patients with PD and depression. In addition, beneficial effects on motor and cognitive symptoms have also been reported.¹⁶⁵

DBS has emerged as an important treatment option for PD, particularly in patients with advanced disease or with side effects that limit the usefulness of dopaminergic medications. The most common sites for stimulator placement in PD are the subthalamic nucleus (STN) and the globus pallidus interna (GPi). The effects of DBS for PD are most clearly seen in the improvement of motor symptoms, but changes in mood have been observed as well. The evidence has been mixed, with some early reports of increased incidence of depression following DBS surgery and subsequent larger randomized trials showing little difference in mood outcomes of DBS versus medical management of PD.¹⁶⁶ Although some reports have suggested greater risk of depression following DBS placement in the STN as compared with GPi, a study designed to compare these sites using mood and cognitive measures as primary endpoints failed to find a significant difference.¹⁶⁷ Stimulation of STN often allows for reduction of dopaminergic medication due to less prominent motor symptoms, but dose reduction of levodopa or dopamine agonists may exacerbate underlying depressive symptoms which these medications had been treating as well. Although premorbid well-controlled depression is not a contraindication to DBS, surgery can be associated with a relapse of depression, and all DBS patients should be monitored postsurgically for emergence or evolution of depressive symptoms.¹⁶⁸

PSYCHOSOCIAL TREATMENT

Nonpharmacological treatments appear useful for depression in PD as well. A randomized controlled trial of CBT for depression in PD adapts CBT to the unique needs of patients with PD to include exercise, behavioral activation, thought monitoring and restructuring, relaxation training, worry control, and sleep hygiene. CBT for patients with PD was supplemented with individual caregiver sessions designed to equip caregivers with the skills necessary to facilitate practice of CBT at home.¹⁶⁹ In comparison with the control group, which received clinical monitoring alone, the treatment group shows significant improvement in measures of depression. Further, this study reveals that caregiver participation rather than patient factors (motor disability, psychiatric comorbidity, and executive function) predicts treatment response.¹⁷⁰

There is some evidence for group therapy in the treatment of depression in PD. A small-scale randomized waitlist controlled trial of group therapy demonstrated a significant reduction in depression in group participants; the group therapy used psychodrama methods (e.g., role play; emotional expression) in 12 sessions, also providing education on Parkinson’s and information on coping skills and adaptive resources.¹⁷¹ Patient Education Program Parkinson (PEPP) which is a standardized program using CBT techniques administered to small groups of patients and caregivers also shows that mood improves significantly for the treatment group.¹⁷²

A review of psychosocial treatments for depression and anxiety in PD points out that interventions that primarily target symptoms of PD have secondary mood benefits.¹⁷³ A study of multidisciplinary rehabilitation including individually administered physical, occupational, and speech therapies, group relaxation exercises, expert lectures, and caregiver groups showed significant improvement in mood, although the benefits were not sustained 6 months after termination.^174,175 This suggests the need for longer-term interventions or maintenance sessions to address the continuing needs of patients and the progressive nature of PD itself.

SUMMARY

PD is a neurodegenerative illness whose core pathology is loss of dopaminergic neurons in the nigrostriatal pathway. The reduction in dopaminergic tone in the ventral striatum is associated with emergence of depression in patients with PD. The diagnosis of, and the progressive symptoms and disability from, a neurodegenerative process also can be psychologically destabilizing and trigger depression symptoms or a major depressive episode. Depression, therefore, is prevalent among patients with PD and can lead to significant additional morbidity. Recognition of depression in PD can be challenging, due to the overlap of neurovegetative symptoms of depression and motor symptoms of PD. Because the effects of depression can exacerbate functional deterioration of patients with PD, routine screening of this population for depression is important. Multiple treatment modalities are beneficial, including TCAs, SSRIs, SNRIs, bupropion, and nonpharmacological treatments such as individual and group CBT for both patients and caregivers. Treatment of PD with levodopa and dopamine agonists will often ameliorate depression. In addition, MAO-B inhibitors, such as rasagiline, can be used to treat both depression and motor symptoms in patients with both PD and depression. ECT should be considered for severe or medication-refractory depression (see box 5-10 for a review of Anxiety in Parkinson’s Disease).

EPIDEMIOLOGY

Huntington disease (HD) is a genetic, autosomal dominant, progressive, neurodegenerative disease with motor and psychiatric features. Although the movement disorder associated with HD has been considered its core feature and the disease was previously known as Huntington chorea,¹⁷⁶ its psychiatric symptoms have been observed and reported since its initial description.¹⁷⁷ Psychiatric disturbances, including depression, anxiety, mania, and psychosis are now considered one of the triad of core symptoms of HD, along with movement disorder and cognitive decline. HD is progressive, with a uniformly poor prognosis. Mean survival after diagnosis ranges from 10 to 15 years, typically with an extended period of severe disability and loss of independent function.

The prevalence of HD varies by ethnicity and geographic location. It is estimated to occur in 5.7 per 100,000 in people of Northern European ancestry, and is less prevalent in populations with non-European roots.¹⁷⁷ Diagnosis is most often made after the emergence of motor symptoms, although genetic testing can identify the disease before any symptoms are apparent. The movement symptoms of HD typically have onset in the fourth or fifth decade, although earlier and even juvenile onset cases are seen. There is some evidence that psychiatric symptoms, particularly depression, and cognitive impairment can precede motor symptoms for years. Most attempts to quantify the length of prodromal mood symptoms are hampered by recall bias and a retrospective approach. A cross-sectional study comparing 55 HD gene carriers without motor symptoms, 85 gene carriers with motor symptoms, and 56 noncarrier first-degree relatives of HD carriers with the general population¹⁷⁸ shows that gene carriers, both symptomatic and presymptomatic, have a significantly increased prevalence of MDD. The evidence for depression in clinically evident HD is unequivocal (Fig. 5-1). MDD affects up to 40% of patients, increasing up to 60% when subsyndromal depression is included.¹⁷⁹

PATHOPHYSIOLOGY

HD¹⁸⁰ results from a mutation on the short arm of chromosome 4, in a gene for the protein product huntingtin. This gene contains a series of CAG repeats of variable length, whereas the normal allele contains 10 to 35 repeats. Alleles carrying 40 or more repeats are considered positive for the mutation, whereas carriers with 36 to 39 repeats are considered “indeterminate” and may show some symptoms due to incomplete penetrance, or remain asymptomatic. As in other trinucleotide repeat disorders, an increased number of repeats does not clearly correlate with severity of symptoms but does correlate with earlier age of onset.

The role of the huntingtin protein in normal neuronal function is unclear, but the abnormal protein forms intranuclear and cytosolic inclusions. This leads to impaired neuronal function and cell death.¹⁸¹ The striatum is preferentially affected and structural imaging shows caudate atrophy with increasing severity as the disease progresses. The damage to the striatum causes the movement symptoms of HD, and likely also contributes to the depressive symptoms. The caudate has connections with multiple frontal-subcortical circuits, important for emotional regulation. The medial caudate has rich limbic connections as well. Downstream effects from damage to the caudate affect function of orbitofrontal and prefrontal cortices, demonstrated by reduced glucose metabolism in these areas in depressed patients, including depressed patients with HD.¹⁸² Evidence that the pathophysiology is itself a cause of depression, rather than a reactive or adjustment response to a neurodegenerative disorder, is demonstrated by the increased prevalence of depression in HD as compared with AD.¹⁸³

CLINICAL PRESENTATION

Depression and other psychiatric symptoms often predate the emergence of motor symptoms, and the diagnosis of HD. In a prodromal analysis, 42% of presymptomatic carriers were taking psychiatric medication at the time of the study, compared to 5% of noncarrier relatives.¹⁶⁸ Despite the elevated index of suspicion for depression among clinicians treating patients with HD, depression may still be challenging to diagnose. Bradyphrenia, apathy, and constitutional symptoms such as sleep disturbance and weight loss are features of HD seen even in the absence of depression.

Other psychiatric symptoms can emerge in HD, sometimes in addition to depression. Anxiety is common, notably obsessive-compulsive disorder.¹⁸⁴ Psychosis, primarily paranoia or delusions and less commonly hallucinations, is seen at times, with up to 11% reporting at least one psychotic symptom during the course of illness.¹⁸⁵ Irritability is a common feature early in the course of HD, and this can progress along with motor and cognitive symptoms. Impaired impulse control and behavioral disinhibition can also occur.